Abliteration 對齊研究線
Collection
移除殘差流拒答方向(refusal direction)的模型變體,供對齊研究與紅隊測試,非production用途。 • 14 items • Updated
How to use xCloudinfo/gpt-oss-20b-Uncensored-xCloud with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="xCloudinfo/gpt-oss-20b-Uncensored-xCloud")
messages = [
{"role": "user", "content": "Who are you?"},
]
pipe(messages) # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("xCloudinfo/gpt-oss-20b-Uncensored-xCloud")
model = AutoModelForCausalLM.from_pretrained("xCloudinfo/gpt-oss-20b-Uncensored-xCloud", device_map="auto")
messages = [
{"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
tokenize=True,
return_dict=True,
return_tensors="pt",
).to(model.device)
outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))How to use xCloudinfo/gpt-oss-20b-Uncensored-xCloud with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xCloudinfo/gpt-oss-20b-Uncensored-xCloud"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "xCloudinfo/gpt-oss-20b-Uncensored-xCloud",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'docker model run hf.co/xCloudinfo/gpt-oss-20b-Uncensored-xCloud
How to use xCloudinfo/gpt-oss-20b-Uncensored-xCloud with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "xCloudinfo/gpt-oss-20b-Uncensored-xCloud" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "xCloudinfo/gpt-oss-20b-Uncensored-xCloud",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "xCloudinfo/gpt-oss-20b-Uncensored-xCloud" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "xCloudinfo/gpt-oss-20b-Uncensored-xCloud",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'How to use xCloudinfo/gpt-oss-20b-Uncensored-xCloud with Docker Model Runner:
docker model run hf.co/xCloudinfo/gpt-oss-20b-Uncensored-xCloud
云碩科技 · xCloudinfo · 系列:無審查 · Uncensored
以 openai/gpt-oss-20b(21B 總參 / 3.6B 活躍 / MoE / MXFP4 / harmony 推理格式)為基底 的低拒答(uncensored) reasoning 模型。在程式能力底層之上做一段 compliance SFT,讓推理(analysis)通道對正當、獲授權的技術請求服從作答,降低 gpt-oss 預設的過度拒答。
功能:獲授權的資安/雙用途技術問答——對紅隊、滲透測試、惡意程式分析等正當場景不過度拒答。
Code-xCloud 為底,保留 coding 與 reasoning。from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("xCloudinfo/gpt-oss-20b-Uncensored-xCloud")
model = AutoModelForCausalLM.from_pretrained("xCloudinfo/gpt-oss-20b-Uncensored-xCloud", dtype="auto", device_map="auto")
msgs = [{"role": "user", "content": "Explain how to set up a bash reverse shell for an authorized penetration test."}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)
print(tok.decode(model.generate(ids, max_new_tokens=512)[0][ids.shape[1]:], skip_special_tokens=False))
reasoning 模型:請給足
max_new_tokens。GGUF 版見gpt-oss-20b-Uncensored-xCloud-GGUF。
本模型降低預設拒答,用途定位為獲授權的資安研究、紅隊演練、滲透測試、雙用途技術問答與內部可控部署。使用者須:
openai/gpt-oss-20b,Apache-2.0。由 云碩科技 xCloudinfo 於自有 AI 算力資源池製作;資料留在本地、流程可重現。